martian source errors between networks

martian source errors between networks

Post by Ana Prados Alons » Sat, 21 May 2011 22:11:12



Hello
Some time ago mine logs start to show this message
-------------------
Apr 23 11:03:01 xxxxx kernel: [38048596.800691] martian source
192.168.0.87 from 217.15.41.137, on dev eth1
Apr 23 11:03:01 xxxxx kernel: [38048596.800691] martian source
192.168.0.87 from 217.15.41.137, on dev eth1
Apr 23 11:15:02 xxxxx kernel: [38049523.750307] martian source
192.168.0.87 from 217.15.41.137, on dev eth1
Apr 23 11:15:02 xxxxx kernel: [38049523.750307] martian source
192.168.0.87 from 217.15.41.137, on dev eth1
Apr 23 11:30:03 xxxxx kernel: [38050706.324262] martian source
192.168.0.87 from 217.15.41.137, on dev eth1
Apr 23 11:30:03 xxxxx kernel: [38050706.324262] martian source
192.168.0.87 from 217.15.41.137, on dev eth1
Apr 23 11:49:14 xxxxx kernel: [38052159.170082] martian source
192.168.0.87 from 217.15.41.137, on dev eth1
Apr 23 11:49:14 xxxxx kernel: [38052159.170082] martian source
192.168.0.87 from 217.15.41.137, on dev eth1
-------------------
I'm using debian5 with 2 network card, eth0(217.15.41.137) and eth1
(my local network) 192.168.0.xxx.
This server is used to host virtual machines and I think this is
something related to the openvz

I've been searching in google and I find some thing, but nothing that
I can use to understand what is causing this

I know that martian are usually ignored or at least that what I found,
but I think  in a server with two networks and many virtuals machines,
that is not a good idea. It could mean that something is wrong and I
can't let it go that easily.

If someone know what could be causing this or can give any advice
about it, it would be lovely
Thank you very much

 
 
 

1. martian source ... cant connect to network

I've a very strange problem when trying to connect to a
cable-modem/network with my laptop.

With a ubuntu-based machine the connection is absolutely no problem. I
simply set the correct hw-adress (the modem authenticates the client via
its mac-adress) and then run a dhcp-client and within a second the
connection is established.

When trying the same on my laptop (opensuse 10.0) I dont get a
dhcp-lease but myriads of the following in my kernel-log:

Apr 21 19:31:09 polar kernel: martian source 255.255.255.255 from
10.34.115.65, on dev eth1
Apr 21 19:31:09 polar kernel: ll header:
ff:ff:ff:ff:ff:ff:00:01:5c:22:78:84:08:00
Apr 21 19:31:13 polar kernel: printk: 7 messages suppressed.

Also when I manually assign the correct ip to the interface (same
netmask, broadcast, route as on the working ubuntu-machine) I dont get a
connection but the above messages.

Strange thing is, that the correct ip is a public 84.xx.xx.xx-adress and
not a private 10.x.x.x-adress like in the "martian-source"-logs.

I disabled the firewall but the message is coming directly from kernel.

any idea?

thnx,
peter

2. help with slow ppp connection

3. Networking crashes my PowerMac5400 under PPC linux r4 with "martian code" errors.

4. iptables and dhcp

5. Martian source

6. possible shell script solution

7. martian source 255.255.255.255 from 10.64.39.106, on dev eth0

8. chat-pppd problems?

9. Martian Source ?

10. martian source

11. martian source: 127.0.0.1 on eth0?

12. Martian Source Problem

13. martian source messages in the kernel log